APPLICATION OF AN EXTENSION NEURAL NETWORK METHOD FOR FAULT DIAGNOSIS OF THREE-LEVEL INVERTERS USED IN MARINE VESSELS

Abstract

This study proposes an extension neural network (ENN) algorithm to carry out the fault diagnosis of an inverter in a motor drive system for marine vessels. First, we use PSIM software to construct a three-level neutral point clamped (NPC) inverter, which simulates the occurrence of the fault of any power transistor in the NPC inverter. Simultaneously, we use fast Fourier transforms to transcribe the waveform of phase voltage in the time domain during the fault of any power transistor entering the spectrum in the frequency domain, looking for the corresponding characteristic spectrum during the fault of the power transistor. Next, we establish the relationship between the fault type and the specific characteristic spectrum as the learning basis of the extension neural network. By doing so, we build up a fault diagnosis system for the inverter. Finally, we use the simulation results to verify the feasibility of the proposed inverter fault diagnosis system.